298 TREATMENT OF SPECIFIC ANXIETY DISORDERS
In summary, there has been fairly consistent empirical support from self- report and
biological challenge experiments that panic disorder is characterized by a heightened
sensitivity or perceptual bias to physical sensations, even though they may not have
enhanced physiological reactivity (Ehlers, 1995). The findings of greater perceptual acu-
ity for interoceptive cues (e.g., enhanced cardiac awareness), however, remains uncer-
tain. Moreover, it is clear that contextual factors affect response to physical sensations
and their interpretation. When bodily sensations occur in unexpected or anxious situ-
ations, individuals with panic disorder are likely to be more vigilant and responsive to
changes in their physical or mental state.
Hypothesis 2. Schematic Vulnerability
Panic-prone individuals will endorse more beliefs about the dangerousness of specific
physiological or mental sensations than nonpanic comparison groups.
In their critical review of the cognitive perspective on panic disorder, Roth, Wil-
helm, and Pettit (2005) noted that if individuals with panic disorder did not exhibit
enduring “catastrophic beliefs” when panic attacks are absent, then this would be prob-
lematic for the theory. According to the schema vulnerability hypothesis, individuals
with panic disorder are expected to exhibit stronger endorsement of thoughts, assump-
tions, and beliefs that reflect activation of physiological threat schemas than nonpanic
disorder individuals even in the absence of a panic attack. Unfortunately, very little
research has specifically focused on beliefs in panic disorder. Khawaja and Oei (1992)
developed the 50-item Catastrophic Cognitions Questionnaire to assess misinterpreta-
tions of the dangerousness of specific physical, emotional, and mental states but the
measure failed to differentiate panic from other anxiety disorders (Khawaja, Oei, &
Baglioni, 1994). Greenberg (1989) constructed the 42-item Panic Belief Questionnaire
(PBQ) to assess level of agreement to maladaptive panic- related beliefs. The PBQ had a
moderate correlation with the ASI (r = .55) and panic disorder patients scored higher on
the total score than a social phobia group, although the difference was not statistically
significant (Ball, Otto, Pollack, Uccello, & Rosenbaum, 1995). More recently Wenzel et
al. (2006) reported that the PBQ Physical Catastrophes subscale had strong correlations
with other panic symptom questionnaires and that scores on the measure declined sig-
nificantly with treatment. Inspection of the PBQ item content indicates that only seven
items (17%) tap into beliefs about physical sensations. Thus at present we do not have
a self- report measure that specifically assesses the enduring physiological and mental
threat schemas proposed by the cognitive model.
Currently the strongest self- report evidence for the schema vulnerability hypothesis
comes from research on anxiety sensitivity (see discussion in Chapter 4). Even though
the ASI is not a belief measure per se, it does assess an enduring tendency to interpret
physical sensations in a threatening manner, which is relevant to the nature of preexist-
ing physiological threat schemas. Evidence that individuals with panic disorder score
significantly higher than other anxiety groups, especially on the ASI Physical Concerns
subscale, and that high ASI scores predict response to biological challenge experiments
as well as development of panic attacks is entirely consistent with the schema vulner-
ability hypothesis for panic disorder. However, the same type of research that has been
conducted on the ASI needs to be extended to a specific panic belief measure like the